Incorporating prior knowledge into independent component analysis
    2.
    发明授权
    Incorporating prior knowledge into independent component analysis 有权
    将先前的知识纳入独立成分分析

    公开(公告)号:US08515096B2

    公开(公告)日:2013-08-20

    申请号:US12141103

    申请日:2008-06-18

    IPC分类号: H04B15/00

    CPC分类号: G10L21/0272

    摘要: The quality of sound recorded from a plurality of people speaking at the same time is improved by incorporating prior knowledge into an independent component analysis (ICA) separating algorithm. More particularly, prior knowledge is defined as a probability distribution according to some prior situation (e.g., prior distribution of people in a room). A mixture of sounds (e.g., mixture of voices) from a plurality of sources (e.g., people) captured by one or more recording devices (e.g., microphones) is separated into individual components (e.g., individual voices from respective people) by applying an maximum a posteriori (MAP) ICA algorithm which incorporates prior knowledge of the respective sources (e.g., location of sources) directly into the MAP ICA algorithm thereby allowing recovery of independent underlying sounds associated with individual sources from the mixture. Therefore, incorporating prior knowledge into an ICA algorithm provides sound quality substantially equal to existing ICA systems, but at reduced computational complexity.

    摘要翻译: 通过将先前的知识结合到独立分量分析(ICA)分离算法中,同时发言的多个人记录的声音的质量得到改善。 更具体地,将先验知识定义为根据一些先前情况(例如,在房间中的人的事先分发)的概率分布。 来自由一个或多个记录装置(例如,麦克风)捕获的多个源(例如,人)的声音(例如,混合声音)的混合物通过应用 最大后验(MAP)ICA算法,其将各个源的先前知识(例如,源的位置)直接并入MAP ICA算法,从而允许从混合物恢复与各个源相关联的独立的底层声音。 因此,将现有知识结合到ICA算法中提供了基本上等于现有ICA系统但声音质量降低的计算复杂性。

    INCORPORATING PRIOR KNOWLEDGE INTO INDEPENDENT COMPONENT ANALYSIS
    3.
    发明申请
    INCORPORATING PRIOR KNOWLEDGE INTO INDEPENDENT COMPONENT ANALYSIS 有权
    将先前的知识纳入独立的组件分析

    公开(公告)号:US20090316928A1

    公开(公告)日:2009-12-24

    申请号:US12141103

    申请日:2008-06-18

    IPC分类号: H04B15/00

    CPC分类号: G10L21/0272

    摘要: The quality of sound recorded from a plurality of people speaking at the same time is improved by incorporating prior knowledge into an independent component analysis (ICA) separating algorithm. More particularly, prior knowledge is defined as a probability distribution according to some prior situation (e.g., prior distribution of people in a room). A mixture of sounds (e.g., mixture of voices) from a plurality of sources (e.g., people) captured by one or more recording devices (e.g., microphones) is separated into individual components (e.g., individual voices from respective people) by applying an maximum a posteriori (MAP) ICA algorithm which incorporates prior knowledge of the respective sources (e.g., location of sources) directly into the MAP ICA algorithm thereby allowing recovery of independent underlying sounds associated with individual sources from the mixture. Therefore, incorporating prior knowledge into an ICA algorithm provides sound quality substantially equal to existing ICA systems, but at reduced computational complexity.

    摘要翻译: 通过将先前的知识结合到独立分量分析(ICA)分离算法中,同时发言的多个人记录的声音的质量得到改善。 更具体地,将先验知识定义为根据一些先前情况(例如,在房间中的人的事先分发)的概率分布。 来自由一个或多个记录装置(例如,麦克风)捕获的多个源(例如,人)的声音(例如,混合声音)的混合物通过应用 最大后验(MAP)ICA算法,其将各个源的先前知识(例如,源的位置)直接并入MAP ICA算法,从而允许从混合物恢复与各个源相关联的独立的底层声音。 因此,将现有知识结合到ICA算法中提供了基本上等于现有ICA系统但声音质量降低的计算复杂性。

    Interactive design, synthesis and delivery of 3D character motion data through the web
    4.
    发明授权
    Interactive design, synthesis and delivery of 3D character motion data through the web 有权
    通过网络互动设计,综合和交付3D角色运动数据

    公开(公告)号:US08704832B2

    公开(公告)日:2014-04-22

    申请号:US12370269

    申请日:2009-02-12

    IPC分类号: G06T13/00

    摘要: Systems and methods are described for animating 3D characters using synthetic motion data generated by generative models in response to a high level description of a desired sequence of motion provided by an animator. An animation system is accessible via a server system that utilizes the ability of generative models to generate synthetic motion data across a continuum to enable multiple animators to effectively reuse the same set of previously recorded motion capture data to produce a wide variety of desired animation sequences. An animator can upload a custom model of a 3D character and the synthetic motion data generated by the generative model is retargeted to animate the custom 3D character.

    摘要翻译: 描述了使用由生成模型生成的合成运动数据来响应由动画制作者提供的期望运动序列的高级描述来动画化3D字符的系统和方法。 可以通过服务器系统访问动画系统,该服务器系统利用生成模型的能力来跨连续体生成合成运动数据,以使多个动画师有效地重用同一组先前记录的运动捕捉数据以产生各种所需的动画序列。 动画师可以上传3D角色的自定义模型,并将生成模型生成的合成运动数据重新定位,以使自定义3D角色动画化。

    INTERACTIVE DESIGN, SYNTHESIS AND DELIVERY OF 3D CHARACTER MOTION DATA THROUGH THE WEB
    5.
    发明申请
    INTERACTIVE DESIGN, SYNTHESIS AND DELIVERY OF 3D CHARACTER MOTION DATA THROUGH THE WEB 有权
    互动设计,通过网络的3D特征运动数据的合成和传递

    公开(公告)号:US20100073361A1

    公开(公告)日:2010-03-25

    申请号:US12370269

    申请日:2009-02-12

    IPC分类号: G06T15/00 G06T15/70

    摘要: Systems and methods are described for animating 3D characters using synthetic motion data generated by generative models in response to a high level description of a desired sequence of motion provided by an animator. In a number of embodiments, an animation system is accessible via a server system that utilizes the ability of generative models to generate synthetic motion data across a continuum to enable multiple animators to effectively reuse the same set of previously recorded motion capture data to produce a wide variety of desired animation sequences. In several embodiments, an animator can upload a custom model of a 3D character and the synthetic motion data generated by the generative model is retargeted to animate the custom 3D character. One embodiment of the invention includes a server system configured to communicate with a database containing motion data including repeated sequences of motion, where the differences between the repeated sequences of motion are described using at least one high level characteristic. In addition, the server system is connected to a communication network, the server system is configured to train a generative model using the motion data, the server system is configured to generate a user interface that is accessible via the communication network, the server system is configured to receive a high level description of a desired sequence of motion via the user interface, the server system is configured to use the generative model to generate synthetic motion data based on the high level description of the desired sequence of motion, and wherein the server system is configured to transmit a stream via the communication network including information that can be used to display a 3D character animated using the synthetic motion data.

    摘要翻译: 描述了使用由生成模型生成的合成运动数据来响应由动画制作者提供的期望运动序列的高级描述来动画化3D字符的系统和方法。 在多个实施例中,可以通过服务器系统访问动画系统,该服务器系统利用生成模型的能力来跨越连续体生成合成运动数据,以使多个动画师有效地重用同一组先前记录的运动捕获数据以产生广泛的 各种所需的动画序列。 在几个实施例中,动画师可以上传3D角色的自定义模型,并且将由生成模型生成的合成运动数据重新定向以对自定义3D角色动画化。 本发明的一个实施例包括服务器系统,其被配置为与包含运动数据的数据库进行通信,所述运动数据包括重复的运动序列,其中使用至少一个高电平特性描述重复的运动序列之间的差异。 另外,服务器系统连接到通信网络,服务器系统被配置为使用运动数据训练生成模型,服务器系统被配置为生成可经由通信网络访问的用户界面,服务器系统是 被配置为经由所述用户界面接收对期望运动序列的高级描述,所述服务器系统被配置为使用所述生成模型基于所述期望运动序列的高级描述来生成合成运动数据,并且其中所述服务器 系统被配置为经由通信网络发送包括可以用于显示使用合成运动数据动画的3D角色的信息的流。

    System for monitoring vehicle use
    7.
    发明授权
    System for monitoring vehicle use 有权
    车辆使用监控系统

    公开(公告)号:US08660740B2

    公开(公告)日:2014-02-25

    申请号:US12745227

    申请日:2008-11-27

    IPC分类号: G01M17/00 G06F19/00 G06F7/00

    摘要: A system for monitoring of vehicles and particularly heavy vehicles and their compliance with specific network (e.g. road) access conditions uses vehicle telematics solutions. The system includes an in-vehicle unit (IVU) associated with a vehicle being monitored. The IVU includes a receiver for receiving positioning signals, a processor for processing a time-marked log of vehicle data, a storage element for storing the time-marked log and a first wireless communication element for communicating time marked data to a Service Provider (SP) processing apparatus. One or more Service Providers operate Service Provider (SP) processing apparatus. The SP processing apparatus include a SP wireless communication element for receiving time-marked data from one or more IVUs and a SP processor for processing received data. The SP processor is adapted to compare received data from the time-marked log of a vehicle with one or more vehicle-use conditions applicable to that vehicle, and to generate a non-compliance report where the comparison indicates that non-compliant activity has occurred. A SP storage element stores non-compliance reports and relevant time-marked data.

    摘要翻译: 用于监控车辆,特别是重型车辆及其符合特定网络(例如道路)接入条件的系统使用车辆远程信息处理解决方案。 该系统包括与被监视的车辆相关联的车载单元(IVU)。 IVU包括用于接收定位信号的接收器,用于处理车辆数据的时间标记日志的处理器,用于存储时间标记的日志的存储元件和用于将时间标记数据传送到服务提供商(SP)的第一无线通信元件 )处理装置。 一个或多个服务提供商操作服务提供商(SP)处理设备。 SP处理装置包括用于从一个或多个IVU接收时间标记数据的SP无线通信元件和用于处理接收到的数据的SP处理器。 SP处理器适于将来自车辆的时间标记的日志的接收数据与适用于该车辆的一个或多个车辆使用条件进行比较,并且生成不合规报告,其中比较指示不符合规定的活动已经发生 。 SP存储元件存储不合规报告和相关的时间标记数据。

    Error detection in communication systems
    9.
    发明授权
    Error detection in communication systems 有权
    通信系统中的错误检测

    公开(公告)号:US06907549B2

    公开(公告)日:2005-06-14

    申请号:US10109949

    申请日:2002-03-29

    IPC分类号: H04L29/14 G06F11/00

    CPC分类号: H04L69/40

    摘要: A method for determining the impact on a network of a fault in the network, comprises the steps of constructing a layered topological model of the network according to a predetermined protocol, receiving an indication of an erroneous event at a point in the network, and modelling the impact on the network and its services resulting from the event by navigating the topological model. The modelling step includes applying a priority weighting to an erroneous event to determine a priority order in which a sequence of erroneous events may be attended to. The priority order may be based on cost-effectiveness of carrying out a particular repair. The method may alternatively be used to determine the location of a fault with increased certainty by relating an event with a previous erroneous event. The method is preferably implemented by a programmed computer.

    摘要翻译: 一种用于确定对网络中的故障网络的影响的方法,包括以下步骤:根据预定协议构建网络的分层拓扑模型,在网络中的某一点接收错误事件的指示,以及建模 通过浏览拓扑模型对事件造成的对网络及其服务的影响。 建模步骤包括对错误事件应用优先级加权以确定其中可能关注一系列错误事件的优先顺序。 优先顺序可以基于进行特定维修的成本效益。 替代地,该方法可以用于通过将事件与先前的错误事件相关联来确定故障增加的确定性的位置。 该方法优选地由编程计算机实现。

    Elevated bikeway
    10.
    发明授权
    Elevated bikeway 失效
    升高的比丘

    公开(公告)号:US4649588A

    公开(公告)日:1987-03-17

    申请号:US827217

    申请日:1986-02-05

    申请人: Graham Taylor

    发明人: Graham Taylor

    IPC分类号: E01D6/00 E01D9/00

    CPC分类号: E01D6/00 E01D2101/34

    摘要: An elevated bikeway can be constructed to be extremely lightweight and therefore inexpensive by restricting use to bicycle traffic only and by using aluminum as substantially the only structural material. The bikeway includes two spaced apart trusses supporting horizontal bracing, which supports a deck tread surface.

    摘要翻译: 升高的自行车可以构造成非常轻便,因此通过仅限于使用自行车交通和通过使用铝作为基本上唯一的结构材料而成本低廉。 自行车包括支撑水平支撑的两个间隔开的桁架,其支撑甲板的踩踏面。